Investigate the exact mechanism of GBM tumor development specific to TMZ treatment.

Glioblastoma multiforme (GBM) is an extremely heterogeneous tumor with high recurrence and mortality despite current standard treatment. One of the complexities in treating GBM is its ability to acquire chemoresistance, leading to a recurrent tumor that does not respond to treatment. 

To investigate this, we performed single-cell RNA-sequencing before, during, and after TMZ-based chemotherapy in a PDX model in vivo. We used this data to map the projection of cell gene expression along pseudotime within each condition using a trajectory inference (TI) program called Slingshot.

Using the Slingshot program, we have observed that lineage complexity is highest in the untreated PDX (primary tumor) in vivo, reduced during TMZ therapy, and again increased in the post TMZ recurrent tumor. By analyzing these different cell lineages across pseudotime, we have identified polo-like kinase 1 (PLK1), as well as its downstream components, were upregulated during the end of cell development during TMZ therapy as well as during early cell development after TMZ therapy, suggesting that PLK1 may have an essential role in the acute adaptation of GBM cells during therapy. We used two inhibitors of PLK1 in vitro that both led to a significant decrease in cell viability (p<0.0001) when given 6 or 24 hours before TMZ treatment. We also administered inhibitors simultaneously with TMZ showing no significant change in viability compared to TMZ alone. This suggests some PLK1 inhibition is necessary, and that PLK1 is only required for GBM survival during therapy. 

Overall, our data implicate PLK-1 as a critical component of acutely acquired TMZ-chemoresistance during therapy. Our initial data presents an effective mechanism in how GBM cells develop resistance during TMZ therapy and how inhibition of PLK1 pathways may significantly improve clinical outcomes in GBM.